Does Parallel-Wire Orientation Matter? Geometry, Current and Pattern
Does Parallel-Wire Orientation Matter? Geometry, Current and Pattern
For two closely spaced conductors carrying nearly equal radiation current in a symmetric environment, rotating the separation plane may cause only a small far-field change. Near ground, supports, feedlines and real terminations, “vertical or horizontal” is not automatically a mechanical-only choice.
The practical question was simple: when two parallel wires form a folded or terminated HF radiator, should the spacer hold them one above the other or side by side? The honest answer is conditional. If the two conductors really behave as one electrically small cross-section, orientation can be a second-order effect. If their currents or surroundings differ, the separation plane becomes part of the antenna.
Short answer: small spacing makes orientation effects smaller; it does not create a universal “no effect” zone. Check the actual conductor currents, electrical spacing, height, ground, feed and termination before calling spacer direction irrelevant.
Start With the Centreline—and Then Add the Cross-Section
The centreline of the complete radiator usually sets the dominant polarisation and large-scale pattern. If two parallel wires are close together, carry similar in-phase radiation currents and see a rotationally symmetric environment, their combined field approaches that of one thicker conductor along the same centreline.
That ideal case explains why a conventional symmetric folded dipole can closely reproduce the pattern of a single-wire dipole. It does not mean that every folded, loaded or terminated pair has equal current on both wires, or that the surrounding world is rotationally symmetric.
The two-wire field in one observation direction can be represented schematically as:
E ∝ I1ej k n·d/2 + I2e−j k n·d/2d is the separation vector, n the observation direction and k = 2π/λ. When I1 = I2 and k|d| is small, the first-order separation term cancels. When current magnitude or phase differs, a first-order orientation term can remain. That is why current equality matters as much as spacing.
There Is No Sharp 0.05λ Boundary
Electromagnetic behaviour changes continuously with spacing. A value such as 0.05λ can be a useful sensitivity checkpoint, but it is not a physical threshold below which orientation becomes unmeasurable. The acceptable spacing depends on the claim:
- A broadside pattern may tolerate a change that matters to a deep null.
- A small feedpoint-impedance shift may be irrelevant to one matching system and troublesome to another.
- A few centimetres of vertical displacement may be negligible high above ground and significant beside lossy soil, roofing or a metal support.
- The same physical spacing becomes electrically larger as frequency rises.
Use s/λ to scale the problem, not to declare it solved. Sweep spacing and orientation in the complete model and look for convergence of the quantities that matter: impedance, conductor currents, efficiency, null depth, azimuth and elevation pattern.
Vertical Separation Changes Two Heights
With one wire above the other, the conductors no longer have the same height over ground. Their image currents, capacitive coupling to soil or roofing, loss and interaction with nearby structures can differ. The centreline height may be unchanged while the individual boundary conditions are not.
That height difference does not automatically produce a dramatic take-off-angle change. It can, however, disturb current equality, shift impedance or fill a null. The effect grows when the antenna is low, the spacing is electrically larger, one wire approaches a support or the ground is lossy and uneven.
Horizontal Separation Changes What Each Wire Sees Sideways
Side-by-side wires have nominally equal height, but one conductor may be closer to a mast, tree, wall, gutter, feedline or termination enclosure. If the environment is asymmetric, rotating the pair simply exchanges one kind of asymmetry for another.
At bends and corners, the separation plane also changes the three-dimensional path. The inner and outer wires can have different lengths and bend radii, and the short connecting sections become part of the current distribution. A drawing that treats the pair as one centreline hides those differences.
A Folded Radiator Carries More Than One Current Component
L. B. Cebik's folded-dipole analysis separates the conductor currents into radiation-current and transmission-line-current components. In a symmetric resonant folded dipole, the summed radiation current closely follows a single dipole, which explains the similar pattern. The individual wire currents are not simply identical copies everywhere.
Move the source, change wire diameters, insert a termination, add a load or make the surroundings unequal, and the current split can change. A terminated folded structure is therefore not guaranteed to share the orientation insensitivity of an ideal centre-fed folded dipole.
Termination does not freeze the pattern: a resistor changes the reflected wave and current distribution. Its value, position, parasitic capacitance, lead geometry and connection to the two conductors all belong in the orientation analysis.
The Feedline Can Dominate a Small Cross-Section Effect
Unequal current on the intended pair can couple to the feedline exterior. The coax or control cable then becomes another conductor with its own height, route and phase. A small spacer-orientation change may appear harmless in an antenna-only model while the installed feedline changes the pattern much more.
A 1:1 choke “right after the UNUN” is not a universal prescription. Transformation and common-mode control are different circuit jobs, and the correct boundary follows the intended current distribution. Model or measure the feedline exterior on both sides of that boundary across every required band.
Spacer Material Is Part of the RF Geometry
Mechanical strength is not the only property that matters. Permittivity, dielectric loss, water absorption, contamination, ultraviolet ageing and the amount of material between the wires can change the local transmission-line component and the completed antenna's impedance. The effect is largest where electric field is concentrated and where wet hardware creates an asymmetric path.
A small, dry, low-loss spacer may be electrically modest. A continuous web, wet rope, coated support or bulky enclosure may not be. Use the material actually installed rather than assuming every non-conductor is invisible at RF.
How to Decide the Orientation
| Condition | Likely design priority | What to verify |
|---|---|---|
| High, symmetric, closely spaced pair | Mechanical stability may dominate | Current equality, impedance and pattern convergence after rotating the separation plane |
| Low over ground or roofing | Keep conductor heights and loss coupling controlled | Both wire currents, ground loss, elevation pattern and impedance |
| Near mast, wall, tree or feedline | Choose the orientation with the cleaner surrounding geometry | Asymmetric coupling, exterior-feedline current and azimuth pattern |
| Deep-null or directional receive design | Protect phase and current balance | Null depth and direction, transformer parasitics and feedline contamination |
| Upper end of a wide frequency range | Treat the physical spacing as electrically larger | Band-by-band current and pattern, not a lowest-band rule |
Model Both Wires, Then Check the Installation
- Represent the real geometry: model both conductors, end connections, bends, feed, termination and meaningful dielectric or metal supports.
- Include the environment: use the installed height, ground, roof, mast and nearby conductors rather than free space alone.
- Include the exterior feedline: an ideal non-radiating transmission-line card cannot reveal coax common-mode current.
- Rotate only the separation plane: keep the centreline and all other variables fixed so the comparison answers the orientation question.
- Inspect wire currents: compare magnitude and phase at facing positions instead of assuming equality from appearance.
- Converge the model: refine segmentation carefully around close parallel wires, short end connections and bends.
- Restore the baseline in the field: compare impedance, exterior current and pattern or SNR with A/B/B/A geometry when the physical installation allows it.
Bottom line: horizontal versus vertical separation can be a minor mechanical choice only after current equality and environmental symmetry make it one. Closely spaced conductors often behave like one thicker radiator in the leading far field, but ground, feedline, supports, termination and unequal current can make the separation plane electrically visible. Follow the installed currents, not a fixed wavelength threshold.
Engineering References
- L. B. Cebik, W4RNL — Unfolding the Story of the Folded Dipole
- L. B. Cebik, W4RNL — Terminated Wide-Band “Folded Dipole”
- L. B. Cebik, W4RNL — Handling Parallel Feedlines
- L. B. Cebik, W4RNL — A Common-Mode Current Picture Show
- L. B. Cebik, W4RNL — NEC-2 Modelling Limits for Close Parallel Wires
- G. J. Burke, Lawrence Livermore National Laboratory — Antenna Modelling With NEC
Mini-FAQ
- Does rotating a closely spaced wire pair always leave the pattern unchanged? No. The leading far field can be similar when currents and surroundings are symmetric, but installed ground, supports, feed and unequal current can expose the orientation.
- Is spacing below 0.05λ automatically negligible? No. Behaviour changes continuously with spacing, and a deep null or low installation can be sensitive to changes that barely move a broad main lobe.
- Does vertical wire separation improve the take-off angle? Not by itself. It gives the two conductors different heights; the resulting current, loss and pattern change depends on the complete installation.
- Do the two wires of a folded antenna always carry equal in-phase current? No. Radiation and transmission-line current components coexist, and feed, termination, wire dimensions and environment can make the individual currents differ.
- Are spacer materials electrically invisible? No. Their permittivity, loss, water absorption and geometry can affect coupling and impedance, especially where electric field is concentrated.
- Where should the common-mode choke go? At the boundary required by the intended installed current distribution, verified over the operating range—not automatically at one box or wavelength distance.